After listening to the TED talk, I have some views about it. Though I believe overall it is a good and reasonable talk, there are aspects I don’t agree.
One is about “The beginners mind”. Apart from the part where I am skeptical that the word “初心” was wrongly translated, I don’t think that having a beginners mind and being able to see many possibilities necessarily contributes to the creative process. These are separate issues when talking about different realms. If the realm is relatively new and you are one of the “pioneers” in the field, sure, be creative! Think about things that no one has thought about before! Because whatever you come up with, you are one of the first people who does that. The act of you participating in this realm is creating experience, and everything you do is a result of creativity. But if the realm is already relatively mature and there are generations before you that have worked on this issue, having a beginners mind, especially when it comes to methodology and approach, will very unlikely be rewarding. In fact, it might greatly hinder your progress and frustrate people working with you.
For example, if you think you came up with an approach to do something, there are generations before you, and guess why they didn’t adopt that approach? Or, how could you be sure that the idea is unprecedented? Take the game development as an example. It is absolutely okay to innovate and be creative about game genres or design elements in games, since on these topics there are still space to work with, and games should all have new elements so that players don’t get tired. Having a beginners mind on these topics I believe is totally fine. But when it comes to the part of coding and implementing the ideas, there is a certain structure you must stick to. Though there might be variations due to specific needs, the overall structure and patterns should be adhered to. Because before you there are countless precedents in this realm, and they already found the best generally applicable ways (and by “best” here it means quantified using rigorous means) to build software. Not sticking to these rules and methods will only cause you time and energy doing something that is greatly unlikely to be rewarding.